JP4859800B2 - Correlation peak detector - Google Patents

Description

  The present invention relates to a correlation peak detection apparatus that detects a correlation peak of a correlation output signal obtained by despreading a spread spectrum signal.

  For example, when applying spread spectrum as a noise countermeasure to a wireless communication apparatus used in a keyless entry system for an automobile, a spread spectrum spread receiver having a configuration suitable for mounting on a portable device having a limited circuit scale is required. . There are various spectrum despreading methods. When the code length is relatively short, a method using a digital matched filter is effective. This is because when a signal is passed through a digital delay element with a built-in polarity tap having the same pattern as the spreading code pattern, a strong peak (pulse) only at the moment when the phase of the spreading code pattern matches the phase of the digital matched filter pattern Is applied, and the principle that only a very small value is detected in other phases is applied. This method requires the same number of delay elements (shift registers) as the code length of the spreading code, but has the advantage that the code phase matching point can be determined automatically (within one code period). When the code length is relatively short, such as 7 or 15, the circuit scale can be reduced.

  The peak output from the digital matched filter is a peak whose polarity is shifted to “+” or “−” according to information “0” and “1” when the signal is information-modulated. . Information can be demodulated by first detecting the position of this peak and then determining the polarity at that position. Since the peak is a portion where the signal intensity is larger than the other portions, a method for detecting the position of the peak is first required. For this purpose, in general, after obtaining the absolute value of the intensity while ignoring the sign of the signal, a portion having a larger value than other portions is detected. The absolute value of the output of the digital matched filter theoretically shows a change as shown in FIG. 8, and shows a larger value in proportion to the code length at the peak position than at the other positions. A small value is shown. Since there is actually some noise, a change as shown in FIG. 9 is shown. A method for detecting the peak position first is necessary, and various methods have been proposed.

  As such a method, conventionally, a correlation value is compared with a predetermined threshold value to determine the presence or absence of a correlation peak (for example, refer to Patent Document 1), or a peak position in one symbol period is determined by absolute value comparison. There was a method (for example, refer to patent documents 2) to do.

JP 2002-185361 A Japanese Patent Laid-Open No. 9-200801

In the spectrum despreading receiver that detects the peak position by the conventional method as described above, when the communication distance becomes long and the signal becomes weak, or when there is an error in the clock of the reference oscillator of the transmitter and the receiver However, there was a problem that communication could not be performed correctly.
For example, in the method using a threshold value, a predetermined value (threshold value A) is used as a threshold value as shown in FIG. 10, and a place where a received signal shows a larger value is specified as a peak. However, this method has a problem that if the communication distance is increased and the signal intensity is reduced as a whole, the signal intensity at the peak position falls below the threshold value, and the peak cannot be detected. That is, the signal is generally accompanied by noise, and it is necessary not to misdetermine this as a peak. The noise may become considerably large depending on the situation, and FIG. 11 shows a case where a portion exceeding the threshold A is generated due to the noise and this is detected as a false peak. In order to avoid such a phenomenon, the threshold value needs to be increased to some extent.

  In FIG. 12, the threshold value is set to a larger value (threshold value B) to prevent the occurrence of such a false peak. However, if the threshold value is set to a large value, a peak can no longer be detected when the communication distance becomes longer and the signal strength becomes weaker. In FIG. 13, the entire signal is only slightly reduced, but a part of the peak cannot be detected. Since the signal strength varies greatly depending on the communication distance, as shown in FIG. 14, even when the signal at the peak position is clearly larger than other portions (good S / N ratio), the signal strength is very weak. Cannot communicate at all. The method using the threshold has such a drawback, but this method has an advantage that it is not influenced by the clock error of the reference oscillator of the transmitter and the receiver. FIG. 15 shows such a case. Since the clock on the transmitter side is faster than the clock on the receiver side, the peak period is shorter than the expected period, but the receiver can detect the peak properly. Yes.

  On the other hand, in the method of detecting the maximum value in one symbol section without using a threshold, a peak can be detected even if the communication distance becomes long and the signal becomes weak. In this method, as shown in FIG. 16, the section is divided for each period (one symbol length) of the code, and a place where the signal shows the maximum value is searched. In this method, since no threshold is used, even if there is noise as shown in FIG. 17, the peak position can be detected without any problem as long as the peak size is maximum in the section. Further, even if the communication distance becomes long and the signal becomes weak as shown in FIG. 18, it is not affected. Even if the signal intensity becomes very small as shown in FIG. 19, the peak position can be detected as long as the S / N ratio is good and the signal is not buried in noise. As described above, the method of detecting the maximum value has an advantage that the communication distance can be increased because the peak can be detected up to the limit where the signal is buried in the noise even when the signal intensity is weak as compared with the method of using the threshold value. On the other hand, this method has the disadvantage that communication is not possible if there is an error in the clocks of the reference oscillators of the transmitter and receiver.

  For example, if the transmitter clock is faster than the receiver clock, the peak period of the signal appearing on the receiver is shorter than expected at the receiver. In this case, as shown in FIG. 20, there may be two peaks in one section. At this time, the relatively smaller peak is ignored and missing. This of course results in poor communication. Conversely, when the clock on the transmitter side is slower than the clock on the receiver side, the peak period is longer than expected on the receiver side. In this case, as shown in FIG. There may be no peak. In this section, a non-peak location such as noise is detected as a peak, resulting in poor communication. Therefore, such a method for detecting the maximum value requires a separate mechanism for matching the clock on the receiver side with the clock on the transmitter side. In general, a mechanism that always corrects the clock on the receiver side using a feedback loop so that the peak position does not move within the section is adopted, but this causes an increase in circuit scale. In particular, when the code length is short, the ratio of this mechanism to the circuit scale becomes a non-negligible size, which is an unacceptable defect such as an increase in power consumption and a reduction in battery life.

  The present invention has been made to solve the above-described problems, and increases the circuit scale even if the signal strength becomes weak and even if there is a clock error in the reference oscillator of the transmitter and the receiver. An object of the present invention is to obtain a correlation peak detection device that can detect a peak without causing it to occur.

The correlation peak detection apparatus according to the present invention sets an inspection section having a length of √2 times the expected correlation peak period of the correlation output signal when passing the spectrum output signal despread in the first-in first-out manner. When the signal intensity at the center of the section is maximum in the inspection section, the signal is determined to be a correlation peak.

The correlation peak detection apparatus of the present invention provides an inspection section having a length √2 times the expected correlation peak period of the correlation output signal, and the signal intensity at the center of the inspection section is maximum in the inspection section. Since the correlation peak is determined, it is possible to detect the peak without increasing the circuit scale even if the signal strength is weak or there is a clock error in the reference oscillator of the transmitter and the receiver.

Embodiment 1 FIG.
1 is a block diagram showing a correlation peak detection apparatus according to Embodiment 1 of the present invention.
The correlation peak detection apparatus shown in the figure is provided in a receiver that receives a spread spectrum signal, and includes a test section setting unit 1, a comparison determination unit 2, and a peak determination unit 3. The configuration for spectrum despreading such as a digital matched filter in the receiver and the configuration of a reference oscillator for generating a reference clock are well known, and are not shown here.

  In the illustrated correlation peak detection apparatus, when the inspection interval setting unit 1 passes the correlation output signal subjected to spectrum despreading in a first-in first-out manner, the inspection interval has a length that is at least one and less than twice the correlation peak period of the correlation output signal. Is a functional unit for setting The comparison / determination unit 2 is a functional unit that compares whether the signal intensity at the center of the examination interval is maximum in the examination interval. The peak determination unit 3 is a functional unit that determines that the signal is a correlation peak when the signal strength determined by the comparison determination unit 2 is maximum.

FIG. 2 is an explanatory diagram illustrating the basic principle of the correlation peak detection apparatus according to the first embodiment.
In the present embodiment, the correlation output signal to be inspected is passed through the first-in first-out method to the inspection section setting unit 1. For example, in the example of FIG. 2, it may be considered that the position of the examination section is fixed and the signal is moved from left to right. Alternatively, it may be considered that a signal having a periodic peak enters from the left of the inspection section and exits from the right. The length of this inspection section is characterized by a value that is at least one and less than twice the expected peak period. The detection of the peak is determined by comparing the value at the center of the examination section with the value of the other part in the examination section by the comparison determination unit 2 and the peak determination unit 3. That is, the comparison / determination unit 2 compares the absolute value of the signal at the center of the examination section with the value of the other part within the section, and the absolute value of the signal at the center of the examination section is greater than the absolute value of any other part. When it becomes larger, the peak determination unit 3 determines that the position is a correlation peak.

  In this way, it is determined that the correlation peak is detected only when the correlation peak comes to the center of the examination section, and it is not judged that the correlation peak has been detected when the other part comes to the center of the examination section. Only the correlation peak can be selectively extracted. This is shown in FIG. When the correlation peak comes to the center of the inspection section, the correlation peak can be correctly detected as long as the correlation peak is smaller than the correlation peak even if noise is on other portions. Since this inspection method does not use a threshold value, as long as the correlation peak is larger than the noise, the correlation peak can be correctly detected even if the entire signal intensity is reduced. Moreover, there is an error in the reference oscillator clocks of the transmitter and the receiver due to the fact that the length of the inspection interval is more than one time and less than twice the expected peak period, which is a feature of this embodiment. Even if the actual period of the correlation peak is different from that expected, the correlation peak is not overlooked, and a portion that is not a correlation peak is not erroneously recognized as a correlation peak.

  This principle will be described in more detail. The reason why the width of the inspection section needs to be one or more times the expected peak period can be understood by considering the case shown in FIG. In this example, since the width of the inspection section is less than one times the expected peak period, the inspection section may fall between the correlation peaks. At this time, if a noise larger than the other part appears in the center of the inspection section, a false peak is detected. Therefore, in order to prevent such a phenomenon, the length of the inspection section needs to be at least one times the expected peak period. On the other hand, the reason why the width of the inspection section needs to be smaller than twice the expected peak period can be understood by considering the case shown in FIG. In this example, the width of the inspection interval is more than twice the expected peak period, so even if a certain correlation peak comes to the center of the inspection interval, one of the adjacent correlation peaks is larger than that due to the influence of noise. In this case, since the peak is the maximum value in the examination section, the correlation peak to be detected is not detected. Therefore, in order to prevent such a phenomenon, the length of the inspection section needs to be smaller than twice the expected peak period.

  In this way, by setting the length of the inspection section to a value that is at least 1 time and less than 2 times the expected peak period, even if there is an error in the clocks of the reference oscillators of the transmitter and receiver, there is a correlation. Only the peaks can be extracted accurately. For example, if the transmitter clock is faster than the receiver clock, the actual correlation peak period is shorter than expected at the receiver. FIG. 6 shows such a case. In this example, the actual peak interval is shorter than in FIGS. 3 and 5, but since the width of the inspection interval is smaller than twice the expected peak period, the correlation peak interval is slightly shorter. By the way, when the correlation peak comes to the center of the inspection section, the correlation peaks on both sides thereof do not enter the inspection section. For this reason, the situation where the correlation peak is overlooked even though it has come to the center does not occur. Conversely, if the transmitter clock is slower than the receiver clock, the actual correlation peak period is longer than expected on the receiver. FIG. 7 shows such a case. In this example, the actual peak interval is longer than in FIGS. 3 and 4, but since the width of the inspection interval is more than 1 times the expected peak period, the correlation peak interval is slightly longer. By the way, the inspection section does not fall between the correlation peaks. For this reason, the situation where the noise between correlation peaks and a correlation peak is erroneously determined as a peak does not occur.

  Based on the principle as described above, in the correlation peak detection apparatus of the present embodiment, the receiver can correctly extract only the correlation peak from the correlation output after spectrum despreading. That is, even if the signal of the correlation peak becomes weak, the correlation peak can be detected up to the limit where the signal is buried in noise, so that communication can be performed even if the communication distance becomes long and the signal becomes weak. Moreover, even if there is an error in the reference oscillator clock of the transmitter and receiver, there is no need for a mechanism to adjust the clock on the receiver side, the power consumption can be reduced by reducing the circuit scale, and the battery-driven portable device The battery life can be extended.

  As described above, according to the correlation peak detection apparatus of the first embodiment, when the correlation output signal subjected to spectrum despreading is passed through first-in first-out, it is at least 1 and less than 2 times the expected correlation peak period of the correlation output signal. The inspection interval setting unit for setting the inspection interval of the length of, the comparison determination unit for comparing whether the signal intensity at the center of the inspection interval is the maximum in the inspection interval, and the signal intensity determined by the comparison determination unit is the maximum If the signal strength is weak, or even if there is a clock error in the transmitter and receiver reference oscillators, the circuit is equipped with a peak determination unit that determines that the signal is a correlation peak. Peaks can be detected without increasing the scale.

Embodiment 2. FIG.
In the second embodiment, the length of the test section in the test section setting unit 1 is set to √2 times (about 1.4 times) the expected peak period, and the reference oscillators of the transmitter and the receiver are set. The tolerance for clock error is maximized. The rest is the same as in the first embodiment.

  As described in the first embodiment, the length of the inspection section can be arbitrarily selected within a range that is 1 time or more and less than 2 times the expected peak period. The shorter one can reduce the circuit scale. However, in order to maximize the resistance to clock errors, it is effective to set the length of the inspection section to √2.

  That is, the clock error of the reference oscillator of the transmitter and the receiver is generally indeterminate which is faster and which is slower. At this time, if the length of the inspection section is √2 times the expected peak period, the maximum is √ (as viewed from the smallest) regardless of whether the transmitter or receiver clock is faster (larger). It can withstand 2-1 times (about 41%) clock error. This is because the peak interval is 1 / N of the expected peak period when the clock on the transmitter side is N times the clock on the receiver side.

  For example, when the transmitter clock is fast, the correlation peak period is shorter than expected, but when the test interval is √2 times the expected peak period, the transmitter clock is the receiver side. Until the peak of √2 times (about 141%), when the peak comes in the center of the inspection section, the adjacent peak will not enter the inspection section (the distance from the center of the inspection section to the end of the inspection section is √2 / 2 = 1 / √2). Conversely, if the transmitter clock is slow (the receiver clock is fast), the correlation peak period is longer than expected, but the test interval is √2 times the expected peak period. When the clock on the receiver side becomes √2 times (about 141%) of the clock on the transmitter side, the inspection interval does not fall between the correlation peaks (the clock on the receiver side is When the side clock is smaller than √2 times, the transmitter side clock is larger than 1 / √2 when viewed from the receiver side clock, that is, the period is smaller than √2 when viewed from the receiver side.

  As described above, according to the correlation peak detection apparatus of the second embodiment, since the length of the inspection section is set to √2 times the expected correlation peak period, the clock on the transmitter side and the clock on the receiver side are Regardless of which is faster and which is slower, it can tolerate clock deviations up to (√2−1) times when viewed from the smaller clock, so communication is normal even if a reference oscillator with low clock accuracy is used. The apparatus can be manufactured at low cost.

It is a block diagram which shows the correlation peak detection apparatus by Embodiment 1 of this invention. It is explanatory drawing which shows the basic principle of the correlation peak detection apparatus by Embodiment 1 of this invention. It is explanatory drawing which shows operation | movement of the correlation peak detection apparatus by Embodiment 1 of this invention. It is explanatory drawing which shows the reason which needs to be 1 time or more of the peak period of the correlation peak detection apparatus by Embodiment 1 of this invention. It is explanatory drawing which shows the reason why it is necessary to be smaller than 2 times the peak period of the correlation peak detection apparatus by Embodiment 1 of this invention. It is explanatory drawing which shows operation | movement when the clock on the transmitter side of the correlation peak detection apparatus by Embodiment 1 of this invention is faster than the clock on the receiver side. It is explanatory drawing which shows operation | movement in case the clock on the transmitter side of the correlation peak detection apparatus by Embodiment 1 of this invention is later than the clock on the receiver side. It is explanatory drawing which shows the theoretical change of the absolute value of the output of a digital matched filter. It is explanatory drawing which shows the actual change of the absolute value of the output of a digital matched filter. It is explanatory drawing which shows the method of detecting the conventional correlation peak position using a threshold value. It is explanatory drawing which shows the misdetection in the method of detecting the conventional correlation peak position using a threshold value. It is explanatory drawing at the time of making a threshold value into a big value in the method of detecting the correlation peak position using the threshold value in the past. It is explanatory drawing which shows the misdetection at the time of making a threshold value into a big value in the method of detecting the correlation peak position using the threshold value in the past. It is explanatory drawing which shows the other example of a false detection at the time of making a threshold value into a big value in the method of detecting the correlation peak position using the threshold value in the past. It is explanatory drawing which shows that there is no influence of a clock error in the method of detecting the correlation peak position using the threshold value in the past. It is explanatory drawing which shows the method of detecting the conventional correlation peak position by the maximum value in 1 symbol area. It is explanatory drawing which shows the peak position detection operation | movement in the method of detecting the conventional correlation peak position with the maximum value in 1 symbol area. It is explanatory drawing which shows the peak position detection operation | movement in case the communication distance is long in the method of detecting the conventional correlation peak position with the maximum value in 1 symbol area. It is explanatory drawing which shows the peak position detection operation in case the signal strength in the method of detecting the correlation peak position by the conventional method by the maximum value in 1 symbol area is very small. It is explanatory drawing which shows the false detection when the clock on the transmitter side is faster than the clock on the receiver side in the conventional method for detecting the correlation peak position by the maximum value in one symbol interval. It is explanatory drawing which shows the false detection in case the clock on the transmitter side is later than the clock on the receiver side in the conventional method of detecting the correlation peak position by the maximum value in one symbol section.

Explanation of symbols

  1 inspection section setting unit, 2 comparison determination unit, 3 peak determination unit.

Claims (1)

An inspection interval setting unit for setting an inspection interval having a length of √2 times the expected correlation peak period of the correlation output signal when passing the spectrum despread correlation output signal in a first-in first-out manner;
A comparison / determination unit for comparing whether the signal strength at the center of the examination section is maximum in the examination section;
A correlation peak detection apparatus comprising: a peak determination unit that determines that the signal is a correlation peak when the signal strength determined by the comparison determination unit is maximum.

Images